Single Labeled DNA FIT Probes for Avoiding False-Positive Signaling in the Detection of DNA/RNA in qPCR or Cell Media

Article abstract of DOI:10.1002/cbic.201200397

Oligonucleotide hybridization probes that fluoresce upon binding to complementary nucleic acid targets allow the real-time detection of DNA or RNA in homogeneous solution. The most commonly used probes rely on the distance-dependent interaction between a fluorophore and another label. Such duallabeled oligonucleotides signal the change of the global conformation that accompanies duplex formation. However, undesired nonspecific binding events and/or probe degradation also lead to changes in the label-label distance and, thus, to ambiguities in fluorescence signaling. Herein, we introduce singly labeled DNA probes, DNA FIT probes , that are designed to avoid false-positive signals. A thiazole orange (TO) intercalator dye serves as an artificial base in the DNA probe. The probes show little background because the attachment mode hinders 1) interactions of the TO base in cis with the disordered nucleobases of the single strand, and 2) intercalation of the TO nucleotide with double strands in trans. However, formation of the probe-target duplex enforces stacking and increases the fluorescence of the TO base. We explored open-chain and carbocyclic nucleotides. We show that the incorporation of the TO nucleotides has no effect on the thermal stability of the probe-target complexes. DNA and RNA targets provided up to 12-fold enhancements of the TO emission upon hybridization of DNA FIT probes. Experiments in cell media demonstrated that false-positive signaling was prevented when DNA FIT probes were used. Of note, DNA FIT probes tolerate a wide range of hybridization temperature; this enabled their application in quantitative polymerase chain reactions.

Full text of DOI:10.1002/cbic.201200397

DOI: 10.1002/cbic.201200397
Single Labeled DNA FIT Probes for Avoiding False-Positive
Signaling in the Detection of DNA/RNA in qPCR or Cell
Media
Felix Hçvelmann, Lucas Bethge, and Oliver Seitz*[a]
Oligonucleotide hybridization probes that fluoresce upon bind-
ing to complementary nucleic acid targets allow the real-time
detection of DNA or RNA in homogeneous solution. The most
commonly used probes rely on the distance-dependent inter-
action between a fluorophore and another label. Such dual-
labeled oligonucleotides signal the change of the global con-
formation that accompanies duplex formation. However, un-
desired nonspecific binding events and/or probe degradation
also lead to changes in the label–label distance and, thus, to
ambiguities in fluorescence signaling. Herein, we introduce
singly labeled DNA probes, “DNA FIT probes”, that are de-
signed to avoid false-positive signals. A thiazole orange (TO) in-
tercalator dye serves as an artificial base in the DNA probe.
The probes show little background because the attachment
mode hinders 1) interactions of the “TO base” in cis with the
disordered nucleobases of the single strand, and 2) intercala-
tion of the “TO nucleotide” with double strands in trans. How-
ever, formation of the probe–target duplex enforces stacking
and increases the fluorescence of the TO base. We explored
open-chain and carbocyclic nucleotides. We show that the in-
corporation of the TO nucleotides has no effect on the thermal
stability of the probe–target complexes. DNA and RNA targets
provided up to 12-fold enhancements of the TO emission
upon hybridization of DNA FIT probes. Experiments in cell
media demonstrated that false-positive signaling was prevent-
ed when DNA FIT probes were used. Of note, DNA FIT probes
tolerate a wide range of hybridization temperature; this ena-
bled their application in quantitative polymerase chain reac-
tions.
Introduction
Fluorescent nucleobases and base surrogates provide interest-
ing opportunities.[1,2] Environmentally sensitive fluorophores
have been positioned within the base stack of DNA to probe
the local structure and dynamics of nucleic acids,[3] to detect
DNA–protein interactions, and to construct helical fluorophore
clusters.[4–5] One important aim of these investigations has
been to develop DNA diagnostic probes that experience en-
hancement of fluorescence emission upon hybridization with
complementary DNA or RNA targets.[6] Such probes will be
useful for the real-time detection of DNA or RNA in complex
environments such as PCR mixtures,[7] cell lysates, or even live
cells.[8] The majority of DNA detection methods used in appli-
cations rely on the distance-dependent interaction between
two fluorophores. However, ambiguities in fluorescence signal-
ing can arise when probe degradation and/or unintended
binding of probes to proteins induce changes in fluorophore–
fluorophore distance.[9–10] The use of a single fluorophore
probes should provide a solution to this problem and, in addi-
tion, should confer advantages as far as cost-effectiveness is
concerned. Nevertheless, there are only a few examples where-
in singly labeled probes show the properties required to
reduce background in complex biomolecular environments,
that is, 1) visible-light-excitable fluorophores and 2) substantial
(ꢀ fivefold) increases in fluorescence upon binding of comple-
mentary targets.[11]
nucleic acid (PNA) oligomer is exchanged for a single thiazole
orange (TO).[12] The TO dye belongs to a family of DNA interca-
lators which have low fluorescence in the free form because
twisting motions around the central methine bridge lead to
rapid depletion of the TO excited state.[13] In the DNA-bound
form, twisting is hampered and fluorescence is dramatically en-
hanced. PNA FIT probes have useful properties. These probes
show weak fluorescence in the single-stranded state. Hybridi-
zation with fully complementary DNA or RNA is accompanied
by strong enhancements of fluorescence.[14] Of note, PNA FIT
probes discriminate single-base mismatches under condition
where both matched and single mismatched probe–target
complexes form, because the TO-base surrogate fluoresces
only weakly when the neighboring base pair is mismatched.
We applied the PNA FIT probes in real-time PCR-based geno-
typing and in the live cell imaging of viral mRNA.[15,16] Other
laboratories have confirmed the interesting properties of PNA
FIT probes in RNA hybridization studies and in the imaging of
microRNA or mRNA in living cells.[17]
Though PNA-based probes offer advantages as far as stabili-
ty in biological environments and ease of synthetic modifica-
[a] F. Hçvelmann, Dr. L. Bethge, Prof. Dr. O. Seitz
Institut fꢀr Chemie der Humboldt-Universitꢁt zu Berlin
Brook-Taylor Strasse 2, 12489 Berlin (Germany)
E-mail: oliver.seitz@chemie.hu-berlin.de
Recently, we introduced the so-called FIT (forced intercala-
tion) probes, wherein a canonical nucleobase in a peptide
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201200397.
ChemBioChem 0000, 00, 1 – 11
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

O. Seitz et al.
tions are concerned, there are drawbacks.[18] The known proto-
cols for cell transfection or enzymatic modification cannot be
applied, and PNA synthesis is, currently, more costly than DNA
synthesis. We set out to develop a DNA version of FIT probes.
Herein we describe the synthesis and properties of DNA-based
probes in which a member of the family of the thiazole orange
based intercalator dyes serves as a fluorescent base surrogate.
In contrast to previously published probes from Asseline,[19]
Wagenknecht,[20] and Okamoto,[21] the involvement of flexible
linkers between the TO dye and the DNA scaffold is avoided.
Rather, we present a new cyclic building block (cRib(TO),
Scheme 1), in which the flexibility of the TO–backbone linkage
is significantly reduced. We show that the replacement of
bond in dRib(TO) is extremely labile. The carbocyclic 2’-deoxy-
ribose analogue cRib(TO) was expected to confer stability
required for automated DNA synthesis yet retain the character-
istic of the cyclic scaffold.
Synthesis
The l-serinol-linked TO-nucleotide 7 was previously described
(Scheme 2).[22] The synthesis of the carbocyclic nucleotides 6a
and 6b was commenced from dicyclopentadiene, which, in ac-
cordance with a procedure published by Meier, was converted
in a sequence of benzyloxymethylation and hydroboration
reactions into 4a’-carba-a-(d)-5’,3’-dibenzyl-2’-deoxyribose 1.[23]
The unprotected hydroxy group in 1 was converted to pro-
duce the mesylate 2. The reaction of 2 with neat 4-methyl
quinoline (tenfold excess) at 908C over a period of 40 h fol-
Scheme 1. DNA FIT probes and fluorogenic nucleotides synthesized in this
study.
a canonical nucleobase by the TO dye in DNA FIT probes has
no effect on the thermal stability of probe–target complexes.
The turn-on of TO emission upon hybridization of DNA FIT
probes is observed with both DNA and RNA targets. To dem-
onstrate the usefulness of the DNA-based FIT probes, we ex-
plored applications such as quantitative PCR (qPCR) analysis
and RNA detection in cell lysates.
Results and Discussion
Previous studies on PNA and DNA probes had indicated that
the linker connecting the TO fluorochrome with the scaffold
has a critical influence on the fluorescence properties. For ex-
ample, PNA probes in which TO was tethered at carboxyethy-
lene or carboxypentamethylene groups to aminoethylglycine
failed to show the desired turn-on of fluorescence upon hy-
bridization.[12c] A similar behavior was observed when carboxy-
ethylene rather than carboxymethylene was used to anchor TO
at an l-serinol nucleotide in DNA probes.[22] We assumed that
the hypothesized “dTO-nucleoside” dRib(TO) would provide
the rigidity required to improve the enforced intercalation in
DNA-based FIT probes (Scheme 1). However, the N-glycosidic
Scheme 2. a) MsCl, NEt3, CH2Cl2, 93%; b) i: 4-methyl quinoline, 908C; ii: 3-
methyl-2-(methylthio)benzothiazolium tosylate, NEt3, 36% (2 steps); c) BBr3,
CH2Cl2, quant.; d) DMTrCl, EtNiPr2, pyridine, KPF6, 85%; e) 2-cyanoethyl N,N-
diisopropylchloro phosphoramidites, EtNiPr2, CH2Cl2, quant.
&2
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Avoiding False-Positive Signaling
lowed by addition of 2-methylthiobenzothiazolium tosylate in
the presence of NEt3 yielded 3 as a mixture of diastereomers.
The removal of the benzyl protecting groups proceeded
smoothly when BBr3 was used. The subsequent introduction of
the DMTr protecting group proved challenging because of the
poor solubility of 4 in pyridine, DMF, or acetonitrile. However,
the addition of KPF6 and the resulting exchange of the counter
ion in 4 solved the solubility problem. Under these conditions
the DMTr protection succeeded in 85% yield. After chromato-
graphic separation of the two diastereomers, 5a and 5b iso-
mers were separately converted to the phosphoamidites 6a
and 6b.
tide acRib(TO) conferred a significant stabilization to the
DNA–RNA duplex. We concluded that the introduction of a TO
nucleotide in DNA FIT probes does not significantly affect, if at
all, the affinity for the targeted RNA.
Fluorescence properties
The fluorescence of probes ON1X–ON3X was measured before
and after addition of target RNA. Figure 1 shows fluorescence
The phosphoramidites (6a, 6b, and 7) were dissolved in dry
MeCN and used in automated DNA synthesis.[24] The coupling
time for the TO-containing building blocks was prolonged to
3 min. This resulted in up to 99% coupling yield (Figure S1 in
the Supporting Information). Cleavage was achieved by em-
ploying fast deprotection of dG(dmf) (32% NH3 (aq.), 2 h,
558C).
Duplex stability
Our main interest pertains to the detection of RNA in complex
environments, such as cell lysates and living cells. Therefore,
we investigated the hybridization of DNA FIT probes with RNA
targets. We first assessed the effect of the TO modification on
the thermal stability of probe–target duplexes. In a set of 18-
mer DNA FIT probes directed against different segments of b-
actin (nt 1138–1163) from Madin–Darby canine kidney (MDCK)
cells, the TO nucleotides (6a, 6b, and 7) were placed at central
positions.[25] After addition of the complementary RNA target
(RNA1) the thermal stability was evaluated by means of UV-
monitored melt analysis (Table 1). A comparison with the un-
Table 1. Melting temperatures of DNA FIT probes directed against a seg-
ment of RNA coding for b-actin in MDCK cells.[a]
TM [8C]
Figure 1. A) Fluorescence spectra of ON3X (black: X=Ser(TO), gray: X=
bcRib(TO)) before (dashed) and after (solid) addition of RNA1. B) Fluores-
cence enhancement at 535 nm of ON1X–ON3X upon addition of RNA1.
Conditions: 1 mm probe in PBS (10 mm Na2HPO4, 100 mm NaCl, pH 7) and
5 mm RNA target; when added, 378C, lex =485 nm.
ON Sequence 5’!3’
1X CACAGAGTXCTTGCGCTC
X=Ser(TO) X=bcRib(TO) X=acRib(TO)
60.0
60.5
56.3
58.0
62.0
63.0
59.5
54.0
62.0
66.0
61.3
66.3
1
CACAGAGTACTTGCGCTC
2X ACAGAGTAXTTGCGCTCA
ACAGAGTACTTGCGCTCA
3X CAGAGTACXTGCGCTCAG
CAGAGTACTTGCGCTCAG
2
3
spectra and fluorescence enhancements I/I0 (where I0 and I are
the fluorescence intensities of the single-stranded form and
the double-stranded form, respectively). The a-carbocyclic link-
age exhibited only modest enhancements of TO emission (ꢃ
twofold) upon hybridization. By contrast, the b-carbocylic TO
nucleotide in ON3X provided the highest hybridization-in-
duced intensification of fluorescence (I/I0 =12) among the se-
quences tested. The serinol-linked TO nucleotide proved least
sensitive to the sequence environment and showed a useful
average I/I0 ꢁ7.
[a] Target RNA: CACCCC CUGAGC GCAAGU ACUCUG UGU (RNA1). Condi-
tions: 1 mm probe and 1 mm RNA1 in PBS (10 mm Na2HPO4, 100 mm NaCl,
pH 7), absorption 260 nm.
modified oligonucleotides ON1–ON3 revealed marginal desta-
bilization of the formed duplexes when the open-chain serinol
nucleotide Ser(TO) or the b-configured carbocyclic nucleotide
bcRib(TO) was involved. For these TO placements, disruption
of AT base pairs in ON1X and ON3X led to a somewhat
weaker effect (DTm ꢁꢂ18C) than disruption of a GC base pair
in ON2X (DTm ꢁꢂ38C). Of note, the a-carbocyclic TO-nucleo-
Two conclusions can be drawn from these experiments.
First, there is no correlation between duplex stability and fluo-
rescence responsiveness of DNA FIT probes. Second, though
ChemBioChem 0000, 00, 1 – 11
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
3
&
ÞÞ
These are not the final page numbers!

O. Seitz et al.
the b-carbocyclic TO nucleotide might provide the highest
fluorescence enhancement, it was the serinol-linked TO nucleo-
tide which appeared more robust in this sequence context.
This and its ease of synthesis prompted further investigations
of the serinol-linked TO nucleotide.
in conformation reported by DNA molecular beacons, the
single TO dye senses changes of the local environment.[22] As
a result, the TO nucleotide will not fluoresce upon scission of
the phosphodiester backbone. Strong fluorescence emission
will occur when the TO nucleotide stacks with the base pairs
formed upon probe–target recognition. This binding mode is
not available with proteins. Thus, we assumed that DNA FIT
probes are less vulnerable to false-positive signaling.
We explored the open-chain TO nucleotide within another
sequence context and selected a 27-nucleotide segment of
mRNA coding for neuraminidase (NA) of the H1N1 influenza A/
PR/8 virus (nucleotides 582–608).[26] This target segment was
first studied by Wang et al. by using molecular beacon
probes.[27] A “TO walk” was used to explore the dependence of
the fluorescence properties on the immediate neighbors
(Table 2). A useful fluorescence enhancement (I/I0 ꢀ5) upon
We compared the DNA FIT probe (Figure 2A) and the DNA
molecular beacon designs (Figure 2B) and addressed the two
major sources of high background in complex biological envi-
ronments, that is, 1) nonspecific binding and 2) enzymatic di-
gestion of the probe. The probes were dissolved in phosphate
Table 2. Fluorescence enhancement of DNA FIT probes directed against
mRNA coding for neuraminidase of H1N1 influenza A/PR/8.
[a]
ON
Sequence 5’!3’
I/I0
4
5
6
7
8
GGTTTC Ser(TO) GTTATTATGCCGTTGTATTT
GGTTTCA Ser(TO) TTATTATGCCGTTGTATTT
GGTTTCAG Ser(TO) TATTATGCCGTTGTATTT
GGTTTCAGT Ser(TO) ATTATGCCGTTGTATTT
GGTTTCAGTT Ser(TO) TTATGCCGTTGTATTT
GGTTTCAGTTA Ser(TO) TATGCCGTTGTATTT
GGTTTCAGTTAT Ser(TO) ATGCCGTTGTATTT
GGTTTCAGTTATT Ser(TO) TGCCGTTGTATTT
GGTTTCAGTTATTA Ser(TO) GCCGTTGTATTT
GGTTTCAGTTATTAT Ser(TO) CCGTTGTATTT
GGTTTCAGTTATTATG Ser(TO) CGTTGTATTT
GGTTTCAGTTATTATGC Ser(TO) GTTGTATTT
GGTTTCAGTTATTATGCC Ser(TO) TTGTATTT
GGTTTCAGTTATTATGCCG Ser(TO) TGTATTT
GGTTTCAGTTATTATGCCGT Ser(TO) GTATTT
7
6
12
4
11
5
2
6
1
3
5
9
10
11
12
13
14
15
16
17
18
7
10
12
3
[a] Target: RNA2, AAAUACAACGGCAUAAUAACUGAAACC. Conditions: see
Figure 1.
RNA hybridization was provided by 10 of the 15 single TO-la-
beled probes tested. Four probes showed I/I0 ꢀ10. Probe ON6
proved the most responsive and was found to signal the pres-
ence of the complementary RNA target with a 12-fold en-
hancement of fluorescence. For comparison, RNA hybridization
of the molecular beacon MB1, which was recently employed
to detect influenza mRNA in living cells, was accompanied by
sixfold intensification of fluorescence.[27]
Comparison between DNA FIT probe and DNA molecular
beacon in cell lysate
The use of nucleic acid-based probes in complex biological
samples such as cell lysate or live cells is challenging. Nonspe-
cific binding to DNA- or RNA-binding proteins and/or degrada-
tion by nucleases can increase background. These problems
are pertinent to fluorophore/quencher-labeled DNA probes.
For example, undesired protein binding or degradation of DNA
molecular beacon probes will separate the fluorophore from
the quencher and will cause increases in fluorescence in the
absence of target.[28] By contrast, DNA FIT probes operate by a
different signaling mechanism. Rather than the global changes
Figure 2. Relative fluorescence spectra of A) ON17 and B) MB1 in PBS
(dotted), cell lysate (gray) and after addition of complementary RNA target
(RNA2) in cell lysate (black) for intact probe (left) and after complete diges-
tion (right) with DNase I (0.44 mgmLꢂ1) for 1 h at 378C. Conditions: A),
B) see Figure 1, lex =550 nm for detection of MB1. C) Time-dependent fluo-
rescence of MB1 (gray) and ON16 (black) in PBS after addition of DNase I
(0.05 mgmLꢂ1 at 3 min) and RNA2 (at 31 min). Conditions: 0.1 mm probe in
PBS, 0.5 mm RNA2 when added, 378C, lem =535 nm or lem =580 nm for de-
tection of ON16 or MB1, respectively.
&4
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Avoiding False-Positive Signaling
buffered saline (PBS) or crude lysate from MDCK cells, and the
fluorescence was measured before and after addition of RNA
target. The change from the PBS buffer to cell lysate was ac-
companied by an increase in the fluorescence of the unbound
MB1 probe (Figure 2B). This increase in background resulted
in reduced responsiveness to target addition (I/I0 (cell lysate)=
0.65ꢁI/I0 (PBS buffer)). In contrast, DNA FIT probe ON16 fluor-
esced with lower intensity in cell lysate than in PBS (Fig-
ure 2A). We speculate that collisional quenching by cellular
components or a better solubilization of the TO dye might
contribute to the reduced fluorescence intensity in cell lysate.
Regardless of the mechanism involved, the DNA FIT probe
ON16 remained highly responsive in cell lysate.
Next studied was the behavior of probes after degradation
by nucleases. Both DNA FIT probe ON16 and DNA molecular
beacon probe MB1 were digested by treatment with DNase I
(40 mL, 0.44 mgmLꢂ1, ꢀ2000 Umgꢂ1, 1 h, 378C). The digested
probes were added to phosphate buffer or cell lysate. The
DNA FIT probe exhibited insignificant fluorescence in both
media (Figure 2A, right). It was impossible to distinguish TO
emission from the lysate-only background fluorescence. This
experiment proved that the remnants of probe digestion (e.g.,
TO nucleoside) stained neither cellular DNA nor RNA. As ex-
pected for complete probe digestion, the fluorescence signal
remained unaltered when complementary RNA target was
added. The MB1 probe showed the opposite behavior (Fig-
ure 2B, right). Degradation resulted in false-positive signaling
as evidenced by increases in fluorescence (70% in lysate,
220% in PBS) in the absence of RNA target.
Figure 3. Simultaneous measurement of time-dependent fluorescence of
ON21 and ON3Ser(TO) after addition of NA RNA (RNA2, 7 min) and actin
RNA (RNA1, 14 min). Conditions: 0.1 mm probes in lysate of 105 MDCK cells
in 150 mL lysis buffer (50 mm Tris·HCl, 150 mm NaCl, 0.8% PN-40, pH 7.4) and
0.5 mm RNA target when added, 378C, black: TO channel, lex =515 nm,
lem =535 nm; gray: BO channel, lex =460 nm, lem =480 nm.
synthesized the l-serinol(BO) 8 and the carbocyclic BO nucleo-
tide building blocks 9a and 9b (Scheme 3; synthesis shown in
Schemes S1 and S2) and introduced the BO nucleotide into
DNA probes ON19X–ON21X (Table S1). These probes were
complementary to a segment from the mRNA coding for NA
of the H1N1 influenza A/PR/8 virus.[26]
To assess the vulnerability to nuclease degradation, MB
probe MB1 and DNA FIT probe ON17 were jointly dissolved in
PBS buffer, and a diluted DNase I solution (5 mL, 0.05 mgmLꢂ1
)
was added. The signal for MB fluorescence immediately in-
creased (Figure 2C). The addition of target RNA showed that
the MB beacon lost 90% of its responsiveness after 30 min of
incubation in cell lysate spiked with DNase I, whereas the DNA
FIT probe lost only 20% of its responsiveness, as inferred from
the 5.4-fold increase in TO emission upon addition of RNA
target. From this data we concluded that DNA FIT probes have
higher resistance to nuclease cleavage than DNA MB probes,
possibly because the sequence internal modification hinders
attack by endonucleases.
Scheme 3. Synthesized BO nucleotides.
Dual-color RNA detection
The simultaneous detection of two different RNA targets
would greatly facilitate the analysis of gene expression, for ex-
ample, when one probe serves as a positive control for expres-
sion of a housekeeping gene while the other probe responds
to the expression of the target gene. This requires two spec-
trally resolvable dyes. Previous investigations with PNA probes
had shown that the “FIT concept” extends to other members
of the thiazole orange family of asymmetric cyanine dyes.[29]
The fluorescence spectra of TO and BO dyes are resolvable
(Figure S5). It should thus be possible to selectively monitor
BO fluorescence (lex =460 nm, lem =480 nm) with only little
crosstalk with TO fluorescence (lex =515 nm, lem =535 nm). We
The NA-specific BO probe ON21 was combined in phos-
phate buffer with the actin-specific TO probe ON3Ser(TO).
Control experiments confirmed that the enhancement of BO
emission required the presence of NA RNA (Figure S5). Analo-
gously, enhancements of TO emission only occurred when
actin RNA was added. To assess the hybridization behavior of
the DNA FIT probes in complex biomolecular environments,
we performed kinetic measurements in crude lysate from
MDCK cells (Figure 3). NA RNA (RNA2) was added after 7 min
equilibration. The signal measured in the BO channel increased
almost instantaneously. The negligible response in the TO
channel attested to the selectivity of fluorescence signaling.
ChemBioChem 0000, 00, 1 – 11
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
5
&
ÞÞ
These are not the final page numbers!

O. Seitz et al.
upon hybridization with complementary DNA. A 3’ cap was in-
troduced onto probe ON8 to avoid extension during PCR. For
this purpose, the DNA FIT probe was synthesized on propane-
1,3-diol-premodified controlled-pore glass (CPG; see the Sup-
porting Information). Hybridization of the 3’-hydroxypropyl
modified FIT probe ON8hp with target DNA at 57.58C led to
tenfold enhancement of the TO emission intensity (Figure 4A).
The qPCR experiments were performed with NA-specific pri-
mers, and the fluorescence of the FIT probe was read in the
FAM (fluorescein) channel. Figure 4B shows the amplification
curves for a cDNA dilution series.
The resulting calibration curve (Figure 4B, inset) revealed
a high PCR efficiency (96%) and a linear measuring range over
at least eight orders of magnitude. Control experiments carried
out without cDNA (no template control, NTC) or with cDNA
obtained from non-infected cells showed that the DNA FIT
probe ON8hp avoided false-positive signaling in the PCR assay.
Thus, in contrast to nonspecific stains, such as SYBR Gold (Fig-
ure S6), DNA FIT probes do not detect primer dimers, which
are formed in the late stage of PCR (>30 cycles).
Discussion
This work was motivated by the increasing importance that
has been assigned to studies of RNA expression. It was our
aim to explore whether single fluorophore-labeled DNA FIT
probes enable RNA detection in complex environments such
as cell lysates. The design of DNA FIT probes is different from
other reported single-label hybridization probes. In most
probes, the sensor dye is attached at the periphery of the
probe or by flexible linkers. For example, Wagenknecht and As-
seline have attached TO to internal positions with flexible link-
ers.[19,20,30] A similar tactic has been applied for the labeling in
HyBeacons.[31] Light-Up probes contain TO at a terminal posi-
tion of PNA.[11c] In Okamoto’s ECHO probes and Saito’s base-
discriminating nucleotides, the sensor dye hangs from the
base periphery.[11b] As few of these probes provide significant
(ꢀ fivefold) increases in fluorescence upon hybridization, it re-
mains difficult to detect RNA behind the background of the
plethora of biomolecules present in cell lysate. Interactions of
the sensor dye with other biomolecules or biomolecular aggre-
gates must be avoided, or must not lead to fluorescence en-
hancement. We believe that the attachment mode realized in
DNA FIT probes sterically shields the TO dye from interactions
with non-targets. Indeed, the experiments in cell media dem-
onstrated that DNA FIT probes such as ON16 show no sign of
false-positive signaling.
Figure 4. A) Fluorescence emission of ON8hp in PBS before (dashed) and
after (solid) addition of complementary DNA. Conditions: 1 mm probe in PBS
buffer and 5 mm DNA target when added, 57.58C, lex =485 nm. B) ON8hp in
qPCR analysis of cDNA (100 ngmLꢂ1–10 fgmLꢂ1, black solid lines) obtained
from H1N1 infected MDCK cells; no-template controls (gray solid) and cDNA
(10 ngmLꢂ1, gray dashed) from non-infected cells. The dotted line shows the
threshold value used for the calibration curve (inset).
Nevertheless, the TO signal immediately responded when actin
RNA (RNA1) was added. A small but measurable increase in
the BO signal faded within 3 min. These experiments revealed
1) the high specificity of dual-color RNA detection, and 2) the
high response rates of the DNA FIT probes.
qPCR analysis
Gene expression is commonly analyzed by means of qPCR and
fluorogenic hybridization probes that report on the amplifica-
tion of a target sequence. We assessed whether DNA FIT
probes also enable the detection of DNA produced by PCR.
This would facilitate the comprehensive analysis of gene ex-
pression. We analyzed cDNA that had been obtained through
reverse transcription of mRNA isolated from MDCK cells after
infection with the influenza A/PR/8 virus. The aim was to quan-
tify mRNA coding for NA of the influenza A virus. A highly
responsive DNA FIT probe from the TO walk shown in Table 2
was selected in order to study the fluorescence properties
We explored two TO attachment modes. Because of the ro-
bustness of the fluorescence response and the ease of prepa-
ration, we explored the serinol scaffold in more detail. Tucker
has also used this linkage to anchor anthracene to DNA.[33] The
“anthracene base” significantly perturbed duplex stability and
furnished rather modest fluorescence responsiveness. By con-
trast, the TO base in DNA FIT probes does not affect the stabil-
ity of the probe–target duplex and provides high enhance-
ment of TO emission upon hybridization. The DNA FIT probes
were designed in analogy to the PNA FIT probes. However, the
&6
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Avoiding False-Positive Signaling
with water purified in a Millipore purification device. Column chro-
matography was performed with SDS 60 ACC silica gel and moni-
tored by TLC with Silica Gel 60 F254 plates (Merck). Optical rota-
tions were measured at the sodium d-line with a model 241 polar-
imeter (PerkinElmer) with a 100 mm glass cuvette. [a]D values are
given in units of 10ꢂ1 degcm2 gꢂ1. NMR spectra were recorded with
a DPX 300 or Avance II 500 spectrometer (Brucker). The signals of
the residual protonated solvents (CDCl3, [D6]DMSO, [D6]DMSO/TFA,
and CD3CN) were used as references. Coupling constants are given
in hertz. High resolution mass spectra were measured with a
QStarXL spectrometer (ESI+; Applied Biosystems). Semipreparative
HPLC was carried out on a model 1105 HPLC System (Gilson, Mid-
dleton, WI), and analytical RP-HPLC was carried out on the 1105
HPLC and an ACQUITY UPLC System (Waters, Milford, MA). A UV
detector (l=260 nm) was used for the detection. Semipreparative
separations were carried out by using an XBridge BEH130 C18
column (8 mm, 10ꢁ150 mm, Waters) at a flow rate of 4 mLminꢂ1 at
558C. Analytical HPLC was carried out by using an XBridge C18
column (5 mm, 4.6ꢁ250 mm; Waters) at a flow rate of 1 mLminꢂ1
at 558C, or an Acquity BEH300 C18 column (1.7 mm, 2.1ꢁ100 mm;
Waters) at a flow rate of 0.6 mLminꢂ1 at 558C. As mobile phase,
a binary mixture of A (triethylammonium acetate (TEAA; 0.1m,
pH 7), aq.) and B (acetonitrile) was used. MALDI-TOF mass spectra
were measured on Reflex III (Bruker) and Voyager (Applied Biosys-
tems) systems. As matrix we used a mixture of 2,4,6-trihydroxy-
acetophenone (10 mg in EtOH (1 mL)) and diammonium citrate
(50 mg in water (0.5 mL)).
usage of a DNA scaffold is more challenging because in the
free form TO has affinity for single-stranded and double-strand-
ed DNA as well as for RNA. The steric bulk provided by the ser-
inol-based and carbocylic TO linkages prevents interactions in
cis and helps decrease the fluorescence of unbound probes.
The experiments with cell media showed that DNA FIT
probes can solve specificity problems pertinent to dual-labeled
probes, such as the frequently used DNA molecular beacons.
Probes that draw on two chromophores and the hybridization-
induced change of the global conformation will provide false-
positive signals when the two interacting chromophores are
separated upon degradation by nucleases or unintended bind-
ing to proteins or other biomolecules.[9,10] By contrast, neither
degradation nor exposure to crude cell lysate was a source of
false-positive signaling when DNA FIT probes were used.
Conclusions
The collected data indicate that singly labeled DNA FIT probes,
in which a member of the thiazole orange (TO) family of inter-
calator dyes occupies the position of a canonical nucleobase,
enable fluorescence-based detection of complementary DNA
and RNA in complex environments. We explored two modes of
attachment. Both the linkage by a carbocyclic scaffold and the
linkage by a serinol-based open-chain ribose analogue provid-
ed high stability of the probe–target complexes formed. Hy-
bridization to a nucleic acid target was signaled by up to 12-
fold enhancement of thiazole orange (TO) emission. Experi-
ments in cell lysate revealed that the fluorescence of the un-
bound DNA FIT probes remained low. Neither degradation by
DNase I nor exposure to crude cell lysate was a source of false-
positive signaling. The method extends to other colors. The
combined use of a TO nucleotide (lex =515 nm, lem =535 nm)
and a BO nucleotide (lex =460 nm, lem =480 nm) in two differ-
ent probes enabled the simultaneous, yet fully independent,
detection of two RNA model targets in cellular media. Of note,
the probes tolerated a wide range of hybridization tempera-
tures, which enabled the application to the qPCR analysis of
viral mRNA. A sequence coding for NA of influenza A virus was
targeted in a feasibility study. The resulting calibration curve
revealed a high sensitivity and a linear measuring range over
at least eight orders of magnitude. Thus, the DNA FIT probes
are amongst the few fluorogenic hybridization probes that can
be used for both the detection of RNA in cell media (and prob-
ably also for live-cell RNA imaging) and DNA formed during
a polymerase chain reaction. These properties suggest that
DNA FIT probes will find applications in other endeavors re-
quiring the real time detection of DNA or RNA in complex
samples.
4a’-Carba-a-(d)-5’,3’-dibenzyl-2’-deoxyribofuranosyl-1’-methyl-
sulfonate (2): NEt3 (8.73 g, 86.3 mmol, 12.1 mL) was added to a so-
lution of 1 (3.37 g, 10.8 mmol) in dry CH2Cl2 (200 mL) at 08C under
an argon atmosphere. Subsequently, a solution of methanesulfonyl
chloride (9.89 g, 86.3 mmol, 6.72 mL) in CH2Cl2 (20 mL) was added
drop-wise. After one hour, saturated NaHCO3 (aq.) solution
(100 mL) was added, and the organic phase was separated. The
aqueous layer was extracted with Et2O (100 mL). The combined
organic phases were dried over MgSO4, filtered, concentrated, and
further purified by flash column chromatography (cyclohexane/
EtOAc 10:1, v/v) to yield a colorless oil (3.94 g, 10.1 mmol, 93%).
Rf =0.78 (EtOAc/cyclohexane 1:1, v/v). 1H NMR (300 MHz, CDCl3):
d=1.91 (ddd, J1 =6.1, J2 =8.6, J3 =14.4, 1H; H2’a), 2.13 (m, 1H;
H2’b), 2.25 (m, 1H; H4a’a), 2.38 (m, 1H; H4a’b), 2.57 (m, 1H; H4’),
2.98 (s, 3H; CH3SO2-), 3.47 (m, 2H; H5’a, H5’b), 3.90 (m, 1H; H3’),
4.51 (m, 4H; 2CH2), 5.15 (m, 1H; H1’), 7.27–7.39 ppm (10H, m,
10ArCH); 13C NMR (75 MHz, CDCl3): d=35.1 (C4a’), 38.7 (CH3), 39.2
(C2’), 44.5 (C4’), 70.8 (C5’), 71.4 (CH2), 73.1 (CH2), 80.2 (C3’), 81.8
(C1’), 127.6 (2ArCH), 127.7 (ArCH), 127.7 (ArCH), 127.7 (2ArCH),
128.4 (2ArCH), 128.4 (2ArCH), 138.3 (ArCq), 138.4 ppm (ArCq).
4a’-Carba-a/b-(d)-5’,3’-dibenzyl-2’-deoxyribofuranose(TO)
(3):
Compound 2 (3.94 g, 10.1 mmol) and 4-methyl quinoline (14.5 g,
101 mmol) were heated to 1008C for 48 h. After cooling to room
temperature, CH2Cl2 (200 mL) was added, followed by 3-methyl-2-
thiomethyl-benzothiazolium tosylate (7.42 g, 20.1 mmol) and NEt3
(4.08 g, 40.4 mmol, 5.67 mL). The addition of triethylamine caused
an immediate color change to red. After stirring for 4 h under the
exclusion of light, the mixture was washed four times with HCl
(1m). The organic layer was dried over MgSO4, filtered, and con-
centrated. The residue was further purified by flash column chro-
matography (CH2Cl2/MeOH 90:10, v/v) to yield a red solid (2.10 g,
2.78 mmol, 36%, ratio of diastereomers a/b=1:1.5). Rf =0.53
(CH2Cl2/MeOH/NEt3 90:10:0.1, v/v/v); 1H NMR (300 MHz, CDCl3, 2
diastereomers: d=1.72 (m, 1H), 2.10 (m, 2H), 2.21 (s, 6H; Ts-CH3),
2.31 (m, 2H), 2.53 (m, 5H,), 3.45 (m, 4H; 2H5’a, 2H5’b), 3.77 (m,
Experimental Section
General information: Reagents were purchased from Sigma–Al-
drich, Acros Organics, and TCI-Europe (Zwijndrecht, Belgium).
Solvents CH2Cl2 and THF were dried by using the SPS-800 solvent
purification system (MBraun, Garching, Germany). Dry DMF (H2O<
0.01%) was purchased from Fluka. Aqueous solutions were made
ChemBioChem 0000, 00, 1 – 11
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
7
&
ÞÞ
These are not the final page numbers!

O. Seitz et al.
6H; 2N+CH3), 3.99 (m, 2H; 2H3’), 4.44 (m, 8H; 4Bn-CH2), 4.94 (m,
1H; H1’), 5.10 (m, 1H; H1’), 6.61 (s, 2H; 2-CH=), 7.02–7.62 (m,
40H), 7.88 (m, 4H; 4Ts-H), 8.57 (m, 3H; 3TO-H), 8.80 ppm (m, 1H;
1TO-H); 13C NMR (75 MHz, CDCl3, 2 diastereomers): d=21.2 (2Ts-
CH3), 34.1 (C4a’), 34.2 (2CH3), 34.2 (C4a’), 37.4 (C2’), 38.4 (C2’), 44.5
(C4’), 44.7 (C4’), 60.4 (C1’), 61.5 (C1’), 71.0 (CH2), 71.1 (CH2), 71.5
(C6’), 71.6 (C6’), 73.1 (CH2), 73.2 (CH2), 79.6 (C3’), 81.0 (C3’), 88.5
(-CH=), 88.6 (-CH=), 108.7 (TO-ArCH), 109.5 (TO-ArCH), 111.8 (2TO-
ArCH), 115.9 (1TO-ArCH), 116.9 (1TO-ArCH), 121.9 (1TO-ArCH),
122.0 (1TO-ArCH), 123.9 (1TO-ArCH), 124.0 (1TO-ArCH), 124.2
(1TO-ArCq), 124.3 (1TO-ArCq), 124.5 (2TO-ArC’q), 126.1 (4Ts-ArCH),
126.2 (1TO-ArCH), 126.4 (1TO-ArCH), 126.6 (1TO-ArCH), 126.8
(1TO-ArCH), 127.1 (1TO-ArCH), 127.5 (2TO-ArCH), 127.6 (4Ts-ArCH,
4Bn-ArCH), 127.6 (4Bn-ArCH), 128.1 (TO-ArCH), 128.2 (2Bn-ArCH),
128.3 (2Bn-ArCH), 128.3 (4Bn-ArCH), 128.4 (4Bn-ArCH), 131.1 (1TO-
ArCH), 132.5 (1TO-ArCH), 137.1 (Bn-ArCq), 137.4 (Bn-ArCq) 137.9
(Bn-ArCq), 138.1 (2TO-ArCq), 138.2 (Bn-ArCq), 138.6 (2Ts-ArCH),
139.9 (1TO-ArCq), 140.0 (1TO-ArCq), 140.3 (1TO-ArCH), 140.4 (1TO-
ArCH), 139.9 (2TO-ArCq), 140.0 (2TO-ArCq), 140.4 (1TO-ArCH), 140.3
(1TO-ArCH), 144.7 (2Ts-ArCH), 148.2 (1TO-ArCq), 148.3 (1TO-ArCq),
159.5 (1TO-ArCq), 159.6 ppm (1TO-ArCq); HRMS (ESI): m/z calcd:
585.2570 [C38H37N2O2S]+, found: 585.2568.
H4’), 2.63 (m, 1H; H4a’b), 3.22 (m, 2H; H5’a, H5’b), 3.72 (s, 6H;
2DMT-OCH3), 3.80 (s, 3H; TO-H18), 4.32 (m, 1H; H3’), 5.32 (m, 1H;
H1’), 6.63 (s, 2H; TO-H10), 6:84 (m, 4H; 4DMT-ArH), 7.18 (m, 1H;
TO-H2), 7.23 (m, 1H; DMT-ArH), 7.30 (m, 6H; 6DMT-ArH), 7.34 (m,
1H; TO-H14), 7.43 (m, 3H; TO-H16, 2DMT-ArH), 7.51 (m, 1H; TO-
H15), 7.66 (m, 1H; TO-H6), 7.78 (m, 1H; TO-H13), 7.87 (m, 1H; TO-
H7), 7.98 (m, 1H; TO-H8), 8.27 (m, 1H; TO-H1), 8.48 ppm (m, 1H;
TO-H5); 13C NMR (100 MHz, CD3CN): d=34.5 (TO-H18), 35.2 (C4a’),
41.1 (C2’), 48.0 (C4’), 55.8 (2DMT-OCH3), 61.0 (C1’), 64.5 (C5’), 73.0
(C3’), 86.9 (DMT-Cq), 88.8 (TO-C10), 109.3 (TO-C2), 113.4 (TO-C16),
114.0 (2DMT-ArCH), 114.0 (2DMT-ArCH), 118.7 (TO-C8), 123.4 (TO-
C13), 125.1 (TO-C12), 125.4 (TO-C4), 125.5 (TO-C14), 126.2 (TO-C5),
127.7 (TO-C6), 127.8 (DMT-ArCH), 128.8 (2DMT-ArCH), 129.0 (2DMT-
ArCH), 129.1 (TO-C15), 130.9 (2DMT-ArCH), 130.9 (2DMT-ArCH),
134.1 (TO-C7), 137.1 (DMT-ArCq), 137.2 (DMT-ArCq), 139.0 (TO-C9),
140.2 (TO-C1), 141.4 (1TO-C17), 146.1 (DMT-ArCq), 149.5 (TO-C3),
159.5 (2DMT-ArCq), 161.3 ppm (TO-C11); HRMS (ESI): m/z calcd:
707.2938 [C45H43N2O4S]+, found: 707.2928.
4a’-Carba-a-(d)-5’-DMT-2’-deoxyribofuranose(TO) (5a): 1H NMR
(400 MHz, CD3CN): d=2.05 (m, 1H; H2’a), 2.19 (m, 2H; H4a’a,
H4a’b), 2.39 (m, 1H; H4’), 2.61 (m, 1H; H2’b), 3.18 (m, 2H; H5’a,
H5’b), 3.76 (s, 3H; TO-H18), 3.77 (s, 6H; 2DMT-OCH3), 4.28 (m, 1H;
H3’), 5.19 (m, 1H; H1’), 6.56 (s, 2H; TO-H10), 6:89 (m, 4H; 4DMT-
ArH), 7.21 (m, 1H; TO-H2), 7.27 (m, 2H; DMT-ArH, TO-H14), 7.35
(7H, m, 6DMT-ArH, TO-H16), 7.47 (m, 3H; TO-H15, 2DMT-ArH), 7.61
(m, 1H; TO-H6), 7.71 (m, 1H; TO-H13), 7.78 (m, 1H; TO-H7), 7.84
(m, 1H; TO-H8), 8.42 (m, 1H; TO-H5), 8.57 ppm (m, 1H; TO-H1);
13C NMR (100 MHz, CD3CN): d=34.4 (TO-C18), 35.5 (C4a’), 41.5
(C2’), 48.6 (C4’), 55.9 (2DMT-OCH3), 61.9 (C1’), 65.1 (C5’), 74.3 (C3’),
86.9 (DMT-Cq), 88.6 (TO-C10), 109.3 (TO-C2), 113.3 (TO-C16), 114.0
(4DMT-ArCH), 118.5 (TO-C8), 123.3 (TO-C13), 125.1 (TO-C12), 125.4
(TO-C4), 125.4 (TO-C14), 126.2 (TO-C5), 127.5 (TO-C6), 127.8 (DMT-
ArCH), 128.8 (2DMT-ArCH), 129.0 (2DMT-ArCH), 129.0 (TO-C15),
131.0 (4DMT-ArCH), 133.9 (TO-C7), 137.1 (DMT-ArCq), 137.1 (DMT-
ArCq), 138.9 (TO-C9), 141.3 (1TO-C17), 141.4 (TO-C1), 146.3 (DMT-
ArCq), 149.4 (TO-C3), 159.6 (2DMT-ArCq), 161.1 ppm (TO-C11);
HRMS (ESI): m/z calcd: 707.2938 [C45H43N2O4S]+, found: 707.2930.
4a’-Carba-a/b-(d)-2’-deoxyribofuranose(TO) (4): Under argon,
BBr3 (2.61 g, 1.02 mL, 10.4 mmol) was added dropwise to a cooled
(08C) solution of 3 (1.58 g, 2.09 mmol) in dry CH2Cl2 (60 mL). Com-
plete addition resulted in
a colorless suspension. Saturated
NaHCO3 (aq.) solution (150 mL) was added after 10 min. After
30 min the precipitate was collected, dried under reduced pressure
to give a red powder, which was used without further purification.
Rf =0.15 (CH2Cl2/MeOH/NEt3 90:10:0.1, v/v/v); 1H NMR (300 MHz,
[D6]DMSO, 2 diastereomers in 1:1.5 ratio): d=1.73 (m, 1H), 2.09 (m,
4H), 2.26 (m, 4H) 2.57 (m, 1H), 3.52 (m, 4H; 2H5’a, 2H5’b), 3.96 (s,
6H; 2CH3), 4.12 (m, 2H; 2H3’), 5.45 (m, 2H; 2H1’), 6.82 (s, 2H; 2CH),
7.32 (m, 4H; 4TO-H), 7.53 (m, 2H; 2TO-H), 7.70 (m, 4H; 4TO-H),
7.98 (m, 4H; 4TO-H), 8.21 (m, 2H; 2TO-H), 8.69 (m, 1H; TO-H), 8.75
(m, 2H; TO-H), 8.84 ppm (m, 1H; TO-H); 13C NMR (75 MHz,
[D6]DMSO, 2 diastereomers): d=33.8 (N+-CH3), 33.8 (N+-CH3), 34.0
(2C4a’), 40.2 (C2’), 40.6 (C2’), 48.8 (C4’), 49.3 (C4’), 59.8 (C1’), 60.4
(C1’), 61.8 (C5’), 61.9 (C5’), 70.9 (C3’), 71.9 (C3’), 87.8 (TO-CH=), 87.9
(TO-CH=), 108.1 (2TO-ArCH), 112.8 (1TO-ArCH), 114.0 112.8 (1TO-
ArCH), 117.8 (1TO-ArCH), 117.9 (1TO-ArCH), 122.7 (1TO-ArCH),
122.8 (1TO-ArCH), 123.7 (2TO-ArCq), 124.1 (1TO-ArCq), 124.1 (1TO-
ArCq), 124.2 (1TO-ArCH), 124.3 (1TO-ArCH), 125.7 (2TO-ArCH),
126.6 (1TO-ArCH), 126.6 (1TO-ArCH), 128.0 (2TO-ArCH), 133.0
(1TO-ArCH), 133.1 (1TO-ArCH), 137.5 (1TO-ArCq), 137.7 (1TO-ArCq),
140.1 (1TO-ArCH), 140.2 (2TO-ArCq), 140.9 (1TO-ArCH), 147.9 (1TO-
ArCq), 147.9 (1TO-ArCq), 159.6 (1TO-ArCq), 159.7 ppm (1TO-ArCq);
(ESI): m/z calcd: 405.1631 [C24H25N2O2S]+, found: 405.1628.
Preparation of phosphoramidites: Nucleoside (0.5 mmol) and 2-
cyanoethyl N,N-diisopropylchlorophosphoramidite (1.25 mmol,
296 mg, 279 mL) were added to a solution of freshly distilled
EtNiPr2 (2.5 mmol, 323 mg, 413 mL) in dry CH2Cl2 (10 mL). After 1 h
of stirring at room temperature the reaction was quenched by ad-
dition of saturated NaHCO3 (aq.). The organic layer was separated
and washed twice with saturated NaHCO3 (aq.), dried over MgSO4,
and filtered. The filtrate was collected, and the volatiles were re-
moved at reduced pressure. The crude product was dissolved in
dry CH3CN to a concentration of 0.1m and transferred to an ABI-
style reagent bottle for subsequent use in DNA synthesis.
4a’-Carba-a/b-(d)-5’-DMT-2’-deoxyribofuranose(TO) (5b/a): KPF6
(769 mg, 4.18 mmol) was added to a suspension of compound 4
(1.58 g, 2.09 mmol) in dry pyridine (100 mL) under argon. The
turbid solution became clear, and EtNiPr2 (2.70 g, 20.9 mmol,
3.45 mL) and DMTrCl (3.40 g, 10.0 mmol) were added. After 16 h
the mixture was filtered. The filtrate was concentrated. The residue
was dissolved in CH2Cl2, washed three times with saturated
NaHCO3 (aq.), dried (MgSO4), filtered, and concentrated. The resi-
due was purified by flash column chromatography in order to sep-
arate diastereomers, which were obtained as red solids (5b,
906 mg, 1.06 mmol, 51%; 5a, 604 mg, 0.71 mmol, 34%). Rf =0.80
(CH2Cl2/MeOH/NEt3 90:10:0.1, v/v/v).
DNA synthesis, work-up, and purification: The DNA FIT probes
were assembled by using a model 3400 synthesizer (Applied Bio-
systems) and by the phosphoramidite method (above). CPGs
(1 mmol, pore size 500 ꢂ) were purchased from Proligo (Sigma Al-
drich) and Link Technologies (Bellshill, UK), and DNA synthesis
reagents (dry CH3CN, trichloroacetic acid (3% in CH2Cl2), acetic an-
hydride in 2,6-lutidin/THF (1:1:8), 1-methylimidazole (16% in THF),
and iodine in water/pyridine/THF (3:2:20:75)) were purchased from
Roth or Link Technologies. The phosphoramidites dT, dABz, dCBz,
and dGDMF (0.1m) were used in dry CH3CN. Activator (5-benzylmer-
capto-1H-tetrazole (BMT)) was purchased from emp Biotech (Berlin,
Germany) and used as a 0.25m solution in dry CH3CN. The synthe-
sized phosphoramidites were used at 0.1m in dry CH3CN. The qual-
ity of each coupling step was monitored by measuring the conduc-
4a’-Carba-b-(d)-5’-DMT-2’-deoxyribofuranose(TO) (5b): 1H NMR
(400 MHz, CD3CN): d=1.67 (m, 1H; H4a’a), 2.29 (m, 3H; H2’a, H2’b,
&8
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

Avoiding False-Positive Signaling
tivity of DMT cleavage solutions. The synthesizer was programmed
to yield oligomers carrying the terminal DMT protective group,
“trityl-on”. After synthesis, the resulting CPGs were dried under re-
duced pressure and then transferred to 2 mL Eppendorf tubes.
Aqueous ammonia (1 mL, 32%) was added, and the tubes were
shaken for 2 h at 558C. Subsequently, the tubes were centrifuged
and the supernatant was collected. The volatiles were removed at
reduced pressure, and the samples were dissolved in TEAA buffer
(0.1m, pH 7). The crude product was further purified by RP-HPLC.
Afterwards, the DMTr group was removed by treatment with AcOH
(50% aq.) for 30 min. The reaction mixtures were neutralized with
NEt3, and the crude product was again purified by RP-HPLC. The
resulting oligomers were concentrated to an overall volume of
0.5 mL and desalted by using NAP-5 Sephadex columns (GE
Healthcare). Finally, the oligomers were freeze dried, and the resi-
dues were dissolved in water (Millipore) to a final concentration of
0.1 mm. Identity and purity were determined by using analytical
RP-HPLC or UPLC and MALDI-TOF or ESI(ꢂ)-Q-TOF mass spectros-
copy.
1 pgmLꢂ1 and 100 fgmLꢂ1, 10 fgmLꢂ1). In the no-template control
(NTC), DNA was replaced with water. DNA template (10 ngmLꢂ1
)
obtained from non-infected cells was used as further control. The
solutions were prepared on ice. Each reaction mixture (3ꢁ20 mL
(triple experiments)) was transferred to a well plate, sealed with
a plastic foil, and centrifuged to remove air bubbles. The tempera-
ture protocol of the PCR included initial denaturation (958C,
2 min), followed by 30 cycles of denaturation (958C, 10 s), primer
annealing (57.58C, 30 s), and elongation (728C, 20 s), and finally
cooling to 48C. Fluorescence readout was performed at the anneal-
ing temperature. Target sequence: H1N1A/PR/8 neuraminidase (nt
496–684).[31]
Acknowledgements
Support by a fellowship of the Fonds der Chemischen Industrie
for F.H. is appreciated. We wish to thank Dr. Susann Kummer,
Matthias Schade and Prof. Dr. Andreas Herrmann, Department of
Biology, Humboldt University Berlin, Invalidenstrasse 42, 10115
Berlin (Germany) for providing MDCK cell lysate and cDNA.
Fluorescence spectroscopy: Fluorescence emission spectra were
measured by using a Cary Eclipse (Varian/Agilent Technologies)
with 10 mm quartz cuvettes and phosphate buffer (Na2HPO4
(10 mm, pH 7), NaCl (100 mm)) and background corrected. DNA FIT
probes and target DNA/RNA were added as specified. Spectra were
recorded upon constant emission. After addition of DNA conjugate
(DNA or RNA stock solutions), the samples were heated to 858C
(2 min), and cooled to 378C. Prior to measurement, samples were
allowed to equilibrate for 2 min. For TO conjugates: lex =485 nm,
lem =500–700 nm; for BO conjugates: lex =440 nm, lem =455–
700 nm. The spectra were the averages of five measurement cycles
and blank corrected. For measurements in cell lysate we used
lysate of 105 MDCK cells in lysis buffer (Tris·HCl (50 mm, pH 7.4),
NaCl (150 mm), nonyl phenoxypolyethoxylethanol (detergent;
0.8%)). Degradation experiments were performed with DNase I
(Sigma–Aldrich) at 378C. For complete digestion, the specified
probe (0.675 nmol) was treated with freshly prepared DNase I
(40 mL, 0.44 mgmLꢂ1 in PBS) for 1 h at 378C. Denaturation of
DNase I was achieved by heating to 808C for 15 min. Completion
of digestion was confirmed by addition of target RNA and the
resulting lack of a fluorescence response. Degraded probes were
added to either PBS or cell media. Fluorescence measurements
were carried out at a final concentration of 1 mm in the specified
medium. For time-dependent measurements of fluorescence, TO-
and BO-containing probes were added together in the same cuv-
ette. BO: lex =460 nm, lem =480 nm, TO: lex =485 nm or lex =
515 nm, lem =535 nm; TAMRA: lex =540 nm, lem =580 nm. Neither
samples nor probes were denaturated prior to use. For time-de-
pendent fluorescence with DNase I, probes (0.15 nmol each) were
jointly added to PBS buffer and single-strand fluorescence was
measured for 3 min (I0), followed by addition of freshly prepared
DNase I (10 mL, 0.05 mgmLꢂ1) to a final concentration of 0.1 mm of
each probe (150 mL total volume). After 32 min, target RNA (RNA2,
0.75 nmol) was added.
Keywords: fluorescence · hybridization probes · molecular
beacon · RNA · thiazole orange
[1] E. T. Kool, Acc. Chem. Res. 2002, 35, 936–943.
[2] M. J. Rist, J. P. Marino, Curr. Org. Chem. 2002, 6, 775–793.
[3] a) Y. L. Jiang, J. T. Stivers, Biochemistry 2002, 41, 11236–11247; b) C.
Beuck, I. Singh, A. Bhattacharya, W. Hecker, V. S. Parmar, O. Seitz, E.
Weinhold, Angew. Chem. 2003, 115, 4088–4091; Angew. Chem. Int. Ed.
2003, 42, 3958–3960; c) M. M. Somoza, D. Andreatta, C. J. Murphy, R. S.
Coleman, M. A. Berg, Nucleic Acids Res. 2004, 32, 2494–2507.
[4] R. W. Sinkeldam, N. J. Greco, Y. Tor, Chem. Rev. 2010, 110, 2579–2619.
[5] a) V. L. Malinovskii, D. Wenger, R. Hꢃner, Chem. Soc. Rev. 2010, 39, 410–
422; b) T. J. Bandy, A. Brewer, J. R. Burns, G. Marth, T. Nguyen, E. Stulz,
Chem. Soc. Rev. 2011, 40, 138–148.
[6] a) K. Nakatani, ChemBioChem 2004, 5, 1623–1633; b) R. T. Ranasinghe, T.
Brown, Chem. Commun. 2005, 5487–5502; c) A. P. Silverman, E. T. Kool
in Advances in Clinical Chemistry, Vol. 43 (Ed.: G. S. Makowski), Academic
Press, San Diego, 2007, pp. 79–115.
[7] a) C. A. Foy, H. C. Parkes, Clin. Chem. 2001, 47, 990–1000; b) I. M.
Mackay, K. E. Arden, A. Nitsche, Nucleic Acids Res. 2002, 30, 1292–1305.
[8] a) S. Tyagi, Nat. Methods 2009, 6, 331–338; b) B. A. Armitage, Curr. Opin.
Chem. Biol. 2011, 15, 806–812.
[9] a) A. P. Silverman, E. T. Kool, Trends Biotechnol. 2005, 23, 225–230;
b) C. J. Yang, K. Martinez, H. Lin, W. Tan, J. Am. Chem. Soc. 2006, 128,
9986–9987.
[10] A. K. Chen, M. A. Behlke, A. Tsourkas, Nucleic Acids Res. 2007, 35, e105.
[11] a) T. Ishiguro, J. Saitoh, H. Yawata, M. Otsuka, T. Inoue, Y. Sugiura, Nucleic
Acids Res. 1996, 24, 4992–4997; b) A. Okamoto, K. Tainaka, Y. Ochi, K.
Kanatani, I. Saito, Mol. Biosyst. 2006, 2, 122–127; c) N. Svanvik, A. Stꢄhl-
berg, U. Sehlstedt, R. Sjçback, M. Kubista, Anal. Biochem. 2000, 287,
179–182.
[12] a) O. Seitz, F. Bergmann, D. Heindl, Angew. Chem. 1999, 111, 2340–
2343; Angew. Chem. Int. Ed. 1999, 38, 2203–2206; b) O. Kçhler, O. Seitz,
Chem. Commun. 2003, 2938–2939; c) O. Kçhler, D. V. Jarikote, O. Seitz,
ChemBioChem 2005, 6, 69–77.
[13] a) L. G. Lee, C.-H. Chen, L. A. Chiu, Cytometry 1986, 7, 508–517; b) J.
Nygren, N. Svanvik, M. Kubista, Biopolymers 1998, 46, 39–51; c) V. Karu-
nakaran, J. L. Pꢅrez Lustres, L. Zhao, N. P. Ernsting, O. Seitz, J. Am. Chem.
Soc. 2006, 128, 2954–2962.
qPCR: PCR was performed on a iQ thermocycler (BioRad, Hercules,
CA) with PCR Master Mix S (PEQLAB Biotechnologie, Erlangen, Ger-
many), containing argon-saturated water, reaction buffer high
specificity S, enhancer (solution P), magnesium chloride (2.5 mm),
dNTP-mix long range (200 mm dNTPs), forward primer (CTTGGT
CAGCAA GTGCAT GT, 400 nm; Biotez, Berlin, Germany), reverse
primer (CTCGGG CCATCA GTCATT AT, 400 nm), Taq DNA polymerase
(1 U; PEQLAB Biotechnologie) and of DNA template-solution (1 mL)
[14] D. V. Jarikote, N. Krebs, S. Tannert, B. Rçder, O. Seitz, Chem. Eur. J. 2007,
13, 300–310.
[15] E. Socher, D. V. Jarikote, A. Knoll, L. Rçglin, J. Burmeister, O. Seitz, Anal.
Biochem. 2008, 375, 318–330.
(100 ngmLꢂ1
,
10 ngmLꢂ1
,
1 ngmLꢂ1
,
100 pgmLꢂ1
,
10 pgmLꢂ1
,
ChemBioChem 0000, 00, 1 – 11
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
9
&
ÞÞ
These are not the final page numbers!

O. Seitz et al.
[16] S. Kummer, A. Knoll, E. Socher, L. Bethge, A. Herrmann, O. Seitz, Angew.
Chem. 2011, 123, 1972–1975; Angew. Chem. Int. Ed. 2011, 50, 1931–
1934.
[24] M. H. Caruthers, A. D. Barone, S. L. Beaucage, D. R. Dodds, E. F. Fisher,
L. J. McBride, M. Matteucci, Z. Stabinsky, J.-Y. Tang, Methods Enzymol.
1987, 154, 287–313.
[17] a) A. Tonelli, T. Tedeschi, A. Germini, S. Sforza, R. Corradini, M. C. Medici,
C. Chezzi, R. Marchelli, Mol. Biosyst. 2011, 7, 1684–1692; b) A. G. Torres,
M. M. Fabani, E. Vigorito, D. Williams, N. Al-Obaidi, F. Wojciechowski,
R. H. Hudson, O. Seitz, M. J. Gait, Nucleic Acids Res. 2012, 40, 2152–
2167; c) Y. Kam, A. Rubinstein, A. Nissan, D. Halle, E. Yavin, Mol. Pharm.
2012, 9, 685–693.
[25] MDCK beta-actin; NCBI reference sequence: NM_001195845.1.
[26] H1N1 neuraminidase APR8; GenBank sequence: CY084184.1.
[27] W. Wang, Z.-Q. Cui, H. Han, Z.-P. Zhang, H.-P. Wei, Y.-F. Zhou, Z. Chen,
X.-E. Zhang, Nucleic Acids Res. 2008, 36, 4913–4928.
[28] S. Tyagi, F. R. Kramer, Nat. Biotechnol. 1996, 14, 303–308.
[29] L. Bethge, D. V. Jarikote, O. Seitz, Bioorg. Med. Chem. 2008, 16, 114–125.
[30] C. Holzhauser, H.-A. Wagenknecht, Angew. Chem. 2011, 123, 7406–
7410; Angew. Chem. Int. Ed. 2011, 50, 7268–7272.
[18] P. E. Nielsen, M. Egholm, R. H. Berg, O. Buchardt, Science 1991, 254,
1497–1500.
[19] U. Asseline, M. Chassignol, Y. Aubert, V. Roig, Org. Biomol. Chem. 2006,
4, 1949–1957.
[31] D. J. French, C. L. Archard, T. Brown, D. G. McDowell, Mol. Cell. Probes
2001, 15, 363–374.
[20] S. Berndl, H.-A. Wagenknecht, Angew. Chem. 2009, 121, 2454–2457;
Angew. Chem. Int. Ed. 2009, 48, 2418–2421.
[21] S. Ikeda, T. Kubota, M. Yuki, H. Yanagisawa, S. Tsuruma, A. Okamoto,
Org. Biomol. Chem. 2010, 8, 546–551.
[32] A. Okamoto, Chem. Soc. Rev. 2011, 40, 5815–5828.
[33] N. Moran, D. M. Bassani, J.-P. Desvergne, S. Keiper, P. A. S. Lowden, J. S.
Vyle, J. H. R. Tucker, Chem. Commun. 2006, 5003–5005.
[22] L. Bethge, I. Singh, O. Seitz, Org. Biomol. Chem. 2010, 8, 2439–2448.
[23] S. Jessel, E. Hense, C. Meier, Nucleosides Nucleotides Nucleic Acids 2007,
26, 1181–1184.
Received: June 12, 2012
Published online on && &&, 0000
&10
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 11
ÝÝ
These are not the final page numbers!

FULL PAPERS
FIT for the right track: The fluores-
cence response of DNA probes that
contain a single thiazole orange (TO)
“base” is specific for complementary
RNA and DNA targets. Strong fluores-
cence enhancements are obtained at
varied temperatures and in challenging
media such as cell lysates.
F. Hçvelmann, L. Bethge, O. Seitz*
&& – &&
Single Labeled DNA FIT Probes for
Avoiding False-Positive Signaling in
the Detection of DNA/RNA in qPCR or
Cell Media
ChemBioChem 0000, 00, 1 – 11
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
11
&
ÞÞ
These are not the final page numbers!